2006 ANES Analysis Report

Transcription

1 Defensive Confidence, ANES Analysis Report Dolores Albarracín and Wei Wang Department of Psychology University of Florida History presents abundant examples that people who strongly advocate and defend a given attitudinal position often change this position, becoming converted to points of view that are opposite to the ones they initially held. One reason for such changes is the degree to which individuals perceive that they can defend their attitudes from attack. Ironically, this trait can make them vulnerable to attitude change (Albarracín, 2002). Presumably, people who are confident that their attitudes will survive future challenges are more willing to examine evidence that both supports and contradicts their prior attitudes. In contrast, people who doubt their defensive ability may prefer proattitudinal information over materials that challenge their prior perspectives (see also Byrne, 1961; Olson & Zanna, 1982b; for related views in other domains, see Tesser, 2001). Although denial may in many ways be relatively a primitive defense mechanism, avoiding counter-attitudinal information may preserve the attitudes of people who doubt their defensive abilities. In contrast, individuals who believe that they will effectively self-defend may willingly receive counter-attitudinal information that succeeds in changing their prior attitudes. The present research was concerned with two questions. First, we are interested in determining whether citizens variations in defensive confidence predict polarization of their political attitudes over time. Exposure to counter-attitudinal information may in turn produce change in a direction opposite to the initial attitudes. Although this sequence of events was demonstrated by Albarracin and Mitchell (2004) in the laboratory, it had never been demonstrated in a political context. The ANES Pilot Study presented an ideal framework for this test. During the proposal stage of the ANES Pilot Study, we proposed several items to measure defensive confidence as well as several items measuring exposure and attention to partisan information. Of the proposed items, only one could be included to measure defensive confidence, and none to measure exposure to and attention to partisan information. Despite this limitation, interesting results were obtained. Defensive Confidence Item and Score Distribution Six-hundred and sixty five participants reported their defensive confidence in the Pilot Study from the ANES (American National Election Survey). Of the originally proposed items, one item was selected and revised as follows: If you wanted to defend an opinion of yours, how successfully do you think you could do that? [Extremely successfully, very successfully, moderately successfully, slightly successfully, or not successfully at all? / Not successfully at all, slightly successfully, moderately successfully, very successfully, or extremely successfully?] Only one person provided a don t know response; the majority of respondents (82.1%) manifested that they could very or moderately successfully defend their opinions, as shown by the score distribution below. These findings were encouraging

2 Defensive Confidence, 2 in that the general population respondents sampled for the pilot study could easily answered the question. We also established that there were no gender differences for this item (p <.46). Not successfully at Extremely all Slightly successfully successfully Very successfully Moderately Successfully Figure 1: Distribution of defensive-confidence scores in the pilot study. Correlations with Defensive Confidence Although the key proposed measures of exposure to partisan information could not be included, the pilot study contained measures that were used in exploratory analyses, after refusals to answer and don t know responses were set as missing. In our analyses, we correlated defensive confidence with general exposure to external information, attention to politics, voting behavior, and the display of inconsistencies between party identification and voting (self-identified Republicans voting for the Democratic candidate). Exposure to external information. Prior research (Albarracín and Mitchell, 2004) found that people with high defensive confidence are more willing to receive external information that disconfirms one s attitudes. Although items measuring exposure to partisan information were unfortunately not present in the survey, eight questions tapped exposure to news, as follows: 1. In a typical day, how much time do you spend watching or reading news on internet/printed newspaper/tv or listening news on the radio? 2. In a typical week, how many days do you watch or read news on internet/printed newspaper/tv or listen to news on the radio? Responses to these questions were correlated with defensive confidence yielding the coefficients in Table 1. Of the 8 questions related to exposure to news, only number of days reading news on internet significantly correlated with defensive

3 Defensive Confidence, 3 confidence. The direction implied that people with high defensive confidence spent fewer days reading news on the Internet. The implication of this finding is not clear at this point. Table 1 Correlations between Defensive Confidence and Time of Exposure to News Exposure Internet Print TV Radio Exposure days Internet days Print days TV days Radio days Defensive * confidence * p <.01 Attention to politics. Five items in the ANES pilot study measured respondents attention to politics, as follows Mod14_A1 How interested are you in information about what s going on in government and politics? Mod14_A2 How closely do you pay attention to information about what s going on in government and politics? Mod14_A3 How often do you pay attention to what s going on in government and politics? Mod14_B1 Some people don't pay much attention to political campaigns. How about you? Would you say that you have been VERY MUCH interested, SOMEWHAT interested, or NOT MUCH interested in the political campaigns this year? Mod14_B2 Some people seem to follow what's going on in government and public affairs most of the time, whether there's an election going on or not. Others aren't that interested. Would you say you follow what's going on in government and public affairs most of the time, some of the time, only now and then, or hardly at all? Before correlational analyses, we combined Items A1 and B1 as a measure of interest in politics; and A3 and B2 as a measure of frequency of attention to politics. We then labeled A2 as close attention to politics. As shown in Table 2, all the three indexes had significant positive correlations with defensive confidence. This result implies that people with high defensive confidence seek out information about politics, even when they do not directly show that they seek out counter-attitudinal information. This finding is intriguing and generally supportive of our prior work.

4 Defensive Confidence, 4 Table 2 Correlations between Defensive Confidence and Attention to Politics Interest in Close political attention to information political information Defensive confidence.170*.220*.133* * p <.001. Trust in government. We also correlated defensive confidence with the following items: Frequent attention to political information Mod17_A1 How much of the time do you think you can trust the government in Washington to do what is right? Mod17_A2 How much of the time do you think you can trust the government in State to do what is right? The results appear in Table 3 and suggested positive correlations between trust in government and defensive confidence. Table 3 Correlations between Defensive Confidence and Trust In Government Trust in federal government Trust in state government Defensive confidence.16*.20** * p <.05, ** p <.01. Voting. We also correlated defensive confidence with reports of voting behavior. In this study, three following items were relevant: Mod26_A2 In talking to people about elections, we often find that a lot of people were not able to vote because they weren't registered, they were sick, or they just didn't have time. How about you--did you vote in the elections this November? Mod26_B2 During the past 6 years, did you usually vote in national, state, and local elections, or did you usually NOT vote? Mod26_B3 During the months leading up to the election held on November 7, did you ever plan to vote in that election, or didn't you plan to do that? After recoding Don t know and Refused responses as missing, we calculated the mean of the three items and used it as the index of participants voting

5 Defensive Confidence, 5 behavior. Analyses revealed that voting and defensive confidence correlated significantly (r =.10, p =.05). Party-inconsistent voting (Swinging). The ANES Pilot Study contained an item measuring current partisan identity, which stated As of today, do you think of yourself as [a Republican, a Democrat / a Democrat, a Republican], an Independent, or what? (Mod19_B1). To analyze discrepancies between party identification and reported voting, we computed a variable that indicated whether or not respondents voted in ways consistent with their party identification. This procedure was followed for participants who identified as and voted for either Republican or Democratic candidates. Inconsistencies (change) received a score of 1 and consistencies a score of 0. Then, we analyzed the correlation between these change scores and defensive confidence. The findings from this analysis appear in Table 4. They showed general positive trends, and a significant positive correlation for presidential voting. These numbers imply that as hypothesized, higher defensive confidence led to greater changes from political identification to voting or reverse. Table 4 Correlations between defensive confidence and party-inconsistent voting Defensive confidence * p <.05 President U.S House of local rep. U.S. Senate of local rep. U.S House of national rep..15* U.S. Senate of national rep. Conclusion The ANES Pilot Study provided supported data for our hypothesis that defensive confidence would lead to increased attention to information and potential change in attitudinal positions. The researchers from this team are grateful for this opportunity.

6 Defensive Confidence, 6 References Albarracín, D., & Mitchell, A. L. (2004). The role of defensive confidence in preference for proatttitudinal information: How believing that one is strong can be a weakness. Personality and Social Psychology Bulletin, 30, Byrne, D. (1961). The repression-sensitization scale: Rationale, reliability, and validity. Journal of Personality, 29, Olson, J.M., & Zanna, M.P. (1982b). Repression-sensitization differences in responses to a decision. Journal of Personality, 50,